Enhancing the Sentiment Classification Accuracy of Twitter Data using Machine Learning Algorithms

Sentiment analysis or opinion mining is the study of public opinions, sentiments, attitudes, and emotions expressed in social media. This is one of the most dynamic research areas in natural language processing and text mining in current years. It is a domain that involves the finding of user sentiment, emotion and opinion within natural language text. The growing significance of sentiment analysis coincides with the increase of social media such as reviews, forum discussions, blogs, micro-blogs, Twitter, and social networks. Common applications of sentiment analysis include the automatic determination of whether a review posted online (of a movie, a book, or a consumer product) is positive or negative toward the item being reviewed. This research work shows the various pathways to perform a computational treatment of sentiments and opinions. The main aim of this work is to classify the sentiment of twitter data using machine learning algorithms. The sentiment classifications have been classified into two types which are emotional classification and polarity classification. This work has been carried out on polarity classification, which is used to classify the text such as positive, negative, and neutral. The polarity classification is done by using the subjectivity lexicon. After the polarity classification two machine learning algorithms are employed to enhance the accuracy of sentiment classification. In the Pre-processing phase, the tweets are preprocessed by using various techniques. Sentiment classification is the essential phase, where preprocessed tweets are taken as input to sentiment classification. The sentiment classification can be done by using subjectivity lexicon. The third phase of the proposed work is to compare and evaluate the performance of two machine learning algorithms which are Support Vector Machine and Decision tre